![SpiceLogo1](https://private-user-images.githubusercontent.com/76076246/324143388-8ceb8f09-1c92-4ca5-9448-0b1adccf5666.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.EGibsXkeSC6koNf5UA4XseM71-5SDa-o5PhQyU7oO60)
![SpiceLogo1](https://private-user-images.githubusercontent.com/76076246/324143546-f85cc48e-c685-4e0a-8453-59c55474f185.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.PyGv0r5qvoCTmOCmC8fjc86x75SE2ic_NOLnQX5D0xM)
This research was conducted during the SPICE Summer Program 2023. Supported by the Pacific Intelligence Innovation Initiative, three students were introduced to the power of Natural Language Processing (NLP) through the lense of climate change.
What is NLP?
- The IBM Think definition: Natural language processing, or NLP, combines computational linguistics—rule-based modeling of human language—with statistical and machine learning models to enable computers and digital devices to recognize, understand and generate text and speech.
Students began by reading the Nature paper Computer-assisted classification of contrarian claims about climate change
- Then, students forked the papers github repository to acceess the data and code
- Next, students performed network analysis utilizing the quanteda r package and created word clouds to further understand the linguistical patterns around each of the climate contrarian "super claims"
- After gathering this understanding, students performed sentiment analysis utilizing the EmoRoberta Model from hugging face (the most recent and powerful model in NLP that can detect up to 28 emotions)
- Lastly, the team explored what it's like to train a classification model by diving into the training and testing code behind ClimateBert